An Explainable AI-Driven Multimodal Deep Learning Framework For Intelligent Android Malware Detection

With the rapid growth of Android applications, malware attacks targeting mobile devices have increased significantly, posing serious security and privacy threats to users. Traditional malware detection techniques, including signature-based and rule-based methods, often struggle to identify newly emerging or obfuscated malware variants. To address these limitations, this study proposes an explainable artificial intelligence-based framework, referred to as XAI-Droid, for effective Android malware detection and classification.The proposed system integrates deep learning techniques with explainable AI (XAI) mechanisms to not only improve detection accuracy but also provide transparent and interpretable decision-making. Feature extraction is performed using static analysis techniques, and the processed features are used to train advanced machine learning and deep learning models. To enhance trust and reliability, explanation methods such as feature importance analysis are incorporated to highlight the key attributes influencing classification decisions.Experimental results demonstrate that the proposed framework achieves high detection accuracy while maintaining interpretability, making it suitable for real-world cybersecurity applications. By combining robust classification performance with explainability, XAI-Droid contributes to the development of trustworthy AI-based mobile security systems.